Local Figma MCP Bridge
Unofficial local MCP bridge for reading the currently open Figma file through a Figma plugin. Provides tools to inspect selection, file info, and export selected nodes as SVG/PNG.
README
Local Figma MCP Bridge
English | 日本語
Unofficial local MCP bridge for reading the currently open Figma file through a Figma plugin.
This project is intended for local development and experimentation. It is not affiliated with, endorsed by, or sponsored by Figma.
MCP client
-> local MCP server (stdio)
-> WebSocket bridge (localhost:8787)
-> Figma Plugin UI
-> Figma Plugin main thread
-> currently open Figma file
What It Does
- Checks whether the Figma plugin is connected to the local MCP server.
- Reads basic information about the open Figma file and current page.
- Summarizes the currently selected nodes, including structure, size, position, colors, layout metadata, and child nodes.
- Exports the current selection as
SVGorPNG.
Privacy And Data Handling
This tool can access design data from the Figma file where the plugin is running.
Data that may be sent from Figma to the local MCP server includes:
- File metadata such as the root file name, page name, page id, and Figma file key.
- Selected node metadata such as node ids, names, dimensions, positions, colors, effects, layout properties, and child hierarchy.
- Exported selection contents as base64-encoded
SVGorPNG, which may include images, text, icons, product UI, customer data, or other confidential design material.
The default implementation sends this data only to a local WebSocket server at ws://localhost:8787. It does not send data to a remote service by itself, and the plugin manifest does not allow production network access to external domains.
Before using this with non-public design files, make sure you have permission from the file owner or organization. Do not publish exported design assets, customer work, internal UI, or file-derived sample data without explicit authorization.
Security Notes
- The bridge is designed for local development. Do not expose
FIGMA_BRIDGE_HOSTor port8787to a public network. - Treat any MCP client connected to this bridge as able to request selected Figma node metadata and exports.
- The MCP server does not currently authenticate WebSocket clients. Run it only on a trusted machine and keep the host bound to
localhostunless you have added your own authentication and transport protections. - Do not commit real exported Figma assets,
.envfiles, access tokens, logs, or generated scratch output.
Setup
npm install
npm run build
Load The Figma Plugin
- Open Figma Desktop.
- Choose
Plugins > Development > Import plugin from manifest.... - Select
packages/figma-plugin/dist/manifest.json. - Run
Local Figma MCP Bridge. - The plugin UI will connect automatically when the local MCP server is running.
If the UI shows Disconnected, start the local MCP server and wait for the plugin to reconnect:
npm run dev:mcp
MCP Client Configuration
Any MCP client that supports stdio servers can use this bridge. Add the built MCP server to your MCP client settings. Codex configuration is shown below as one example:
{
"mcpServers": {
"local-figma": {
"command": "node",
"args": [
"/absolute/path/to/packages/mcp-server/dist/index.js"
],
"env": {
"FIGMA_BRIDGE_PORT": "8787"
}
}
}
}
Optional environment variables:
FIGMA_BRIDGE_HOST: WebSocket host. Defaults tolocalhost.FIGMA_BRIDGE_PORT: WebSocket port. Defaults to8787.
MCP Tools
figma_status: Check whether the Figma plugin is connected.figma_file_info: Get basic information about the open Figma file and current page.figma_get_selection: Return a JSON summary of the currently selected Figma nodes.figma_export_selection: Export the current Figma selection as base64-encodedSVGorPNG.
Figma Manifest Network Access
The plugin manifest uses:
{
"networkAccess": {
"allowedDomains": ["none"],
"devAllowedDomains": [
"http://localhost:8787",
"ws://localhost:8787"
]
}
}
This keeps production network access disabled while allowing the local development bridge. If you change this project to send data to remote services, update the manifest, documentation, privacy policy, and user consent flow accordingly.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.